利用反应分子动力学研究煤热解反应中分子与煤颗粒之间的介尺度现象

Understanding the coal pyrolysis process is essential for efficient and clean coal utilization. However, the heterogeneous nature of coal and the complexity of its pyrolysis process involving myriad coupled reaction pathways have made it very difficult to perform unambiguous experiments to study the reaction rates and mechanisms in coal pyrolysis. Particularly, the.devolatilisation reactions in coal pyrolysis are well known to be initiated by free radicals at high temperatures before taking place in an extremely short period of time, and are impractical to detect and replicate in laboratories. Hence it would be useful to explore the initial mechanisms of coal pyrolysis at molecular level with reactive molecular dynamics, a molecular modeling approach that combines MD with an empirical reactive force field (ReaxFF) proposed by van Duin et al. Applications of ReaxFF MD in coal pyrolysis can be found very recently which shown its potential as a useful tool in exploration of reaction mechanism of coal pyrolysis process. Using the new methodology by combining graphics processing unit (GPU)-enabled high performance computing with cheminformatics analysis proposed by the applicant, this project will extend reaction mechanism study of coal pyrolysis and focus on to the molecular coherence evolving of chemical reactions of coal molecules and coal matrix during coal thermal decomposition. The structure evolution of the reaction systems, including the coal molecules, the pyrolyates penetrated by micropores in which diffusion of tar molecules is activated and subject to reaction with the coal matrix with a probability that depends on the relative rates of diffusion and recombination, the intraparticle mass transfer influences on the course of pyrolysis, etc. will be investigated from a perspective of meso-science. The output of the project will be helpful in getting depth understanding of the relationships of structure and reactions and uncover the mechanism on coal pyrolysis process.

认识煤热解过程是实现煤高效和清洁利用的基础和关键。煤本身和热解反应的复杂性导致认识煤热解过程的困难；煤热解被认为是高温自由基过程，因缺乏原位测定技术，限制了对其机理的深入认识。化学反应分子动力学方法(ReaxFF MD)为认识煤热解反应机理提供了新途径，已开始出现的尝试仅涉及煤热解本征化学反应。本项目拟利用ReaxFF MD，进一步将研究延伸到煤热解本征反应与其热解产物向煤基体孔道及其表面扩散传质过程的耦合，基于申请人提出的将GPU并行高性能计算与化学信息学分析相结合的新思路，利用在国际上率先自主研发的高性能计算GPU程序和化学反应分析工具，从介尺度结构角度研究纳米煤基体微反应系统中的由反应介质(煤热解反应演化中的分子)和煤颗粒表界面(微孔道、颗粒表面)构成的介尺度结构的形成机理及其对反应的控制机制，以期系统揭示煤分子结构及其聚集态随热解反应的演化机理，为煤的高效、清洁转化利用提供理论支持

认识煤热解过程是实现煤高效和清洁利用的基础和关键。煤本身和热解反应的复杂性导致认识煤热解过程的困难。化学反应分子动力学方法(ReaxFF MD)为认识煤热解反应机理提供了可能的新途径。本项目基于申请人提出的将GPU并行高性能计算与化学信息学分析相结合研究煤热解的新思路，主要开展煤热解反应分子动力学模拟，研究目标是借助对煤热解过程介尺度结构演变和控制机制的考察，深入认识煤热解反应机理。.本项目构建了多种大规模煤模型(~10,000原子, 尝试建立的最大煤模型为98,000原子)，利用自主研发的基于GPU的高性能计算程序GMD-Reax，对所建立的大规模柳林煤模型进行了ReaxFF MD模拟，其中对近10万原子的模拟尝试是用ReaxFF MD模拟过的最大煤模型。借助自主研发的化学反应分析程序VARxMD，考察了煤热解产物包括中间产物和自由基及其背后的化学反应随温度和时间的演化规律。利用大规模煤分子模型的优势，模拟得到了煤热解各类产物随温度和时间演化图景。.认识煤热解过程中的介尺度现象的关键是体系时空动态演变过程中结构的转折性变化。本项目借助于对木质素热解模拟研究所获得的木质素热解过程中3个阶段产物演化与连接键之间关系的清晰认识，采用相同的模拟策略探究了柳林烟煤热解的三个不同阶段，根据桥键演化行为及其相互竞争的热裂解反应与重聚反应所占比例，可进一步将煤热解第二阶段细分为热解占主导和缩聚开始增长两个子阶段。这两个状态的转折点与小自由基(主要是OH和CH3)的行为转折密切相关，自由基向稳定化方向转变。CH4的生成速率由慢变快的转变有可能作为状态转折的标志物加以检测。.对万原子规模煤模型的ReaxFF MD模拟应用表明，大规模模拟可获得煤热解反应随温度和时间演化的全景式图景，借助于介尺度结构演变和控制机制的探索，获得了对煤热解过程阶段性转折点、相互竞争的热裂解与重聚反应的演化、及其与煤热解产物演化的关系认识。大规模ReaxFF MD模拟结合介尺度科学探索是一种有前景的、研究煤热解的新方法。